This invention relates to a system and method for measuring the rate of application of liquid glue during the manufacture of corrugated board.
In the manufacture of corrugated board a layer of paper referred to as the medium is first passed between two metal, heated and toothed, corrugating rolls that form it into a sinusoidal, fluted shape. A topside layer of paper referred to as the top liner is then bonded with an aqueous, starch-based glue solution to the topside tips of the medium""s flutes. This initial fluting and bonding process is typically referred to as singlefacing, and takes place in an area of the corrugating machine that is typically referred to as the singlefacer.
During singlefacing the glue is first delivered to the singlefacer""s glue pan through which rotates a glue roll. Glue picked up by this glue roll is then metered off by an adjacent metering roll. The gap between this glue roll and metering roll, which is typically referred to as the singlefacer metering roll gap, may be adjusted to produce a specific glue roll film thickness and resulting topside glue application rate. After the glue is applied the medium is held in contact with the top liner for a brief interval by a suitable loading element, such as a pressure roll or loading belt, to produce the bonded, two-layer, singleface web.
The underside of the singleface web""s flutes are then bonded to a second liner, typically referred to as the bottom liner, to form the final three-layer, corrugated board product. This final glue application and bonding process, which is typically referred to as doublefacing, takes place in an area of the corrugating machine that is typically referred to as the glue-machine or glue-unit.
During doublefacing the glue is first delivered to the glue-machine""s glue pan, through which rotates a glue roll. Glue picked up by this glue roll is then metered off by an adjacent metering roll. The gap between this glue roll and metering roll, which is typically referred to as the doublefacer metering roll gap, may be adjusted to produce a specific glue roll film thickness and resulting bottom-side glue application rate. During this bottom-side glue application the medium""s flutes are typically held in contact with the glue roll by an adjacent rider roll or spring-loaded contact bar that is pressed lightly against the top of the singleface web. Then, after the glue is applied to the underside of the medium""s flutes, the medium is subsequently pressed against the incoming bottom liner to initiate the bottom-side bond and complete formation of the three-layered corrugated board.
Some corrugating machines have two or more singlefacers, to form two or more singleface webs, which when bonded together, and to a final bottom liner, permit the formation of a five or more layered corrugated board.
To produce consistent, optimum corrugated board quality, to minimize raw material costs, and to maximize production speeds, glue application rates must be controlled to minimum target levels (across the machine""s width), at both the singlefacer and glue-machine.
Producing well-bonded corrugated board to minimize scrap due to delamination requires consistent, uniform glue application to ensure adequate coverage. Producing flat, pliable corrugated board to minimize scrap due to dimensional warp and cracked score lines, requires optimum board moisture levels that also depend upon consistent glue application rates. The paper used to produce corrugated board typically contains about 6.5% water by weight. The paper in a representative three-layer board with a total raw material weight about 120 lbs/1,000 square feet, thus contains close to 8 lbs of water per 1,000 square feet of board. The minimal amount of glue applied to a board of this weight is about 10 lbs of liquid glue per 1,000 square feet, of which about 25% is solids (starch and additives, such as borax) and 75% water (7.5 lbs/1,000 square feet). An application rate of 50 to 100% higher than this is also not uncommon. As indicated by this representative example the amount of water added to the process with the glue is similar and often greater than that entering the process with the paper. Consequently, effective control of final corrugated board moisture requires not only consistent paper moisture levels, but also control of the glue application rate.
A representative corrugating plant produces about 70 million square feet of board per month. This equates to roughly 50,000 tons of board per year, including over 1,000 tons of starch (solids). In this example about 3,000 tons of water per year, introduced along with the starch, would have to be subsequently evaporated from the board during contact with steam-heated vessels. Even a modest 10% reduction in glue consumption, made possible by the on-line measurement and control of the glue application rate, would reduce this representative corrugating plant""s glue purchases by 100 tons per year, and it""s steam consumption rate by over 300 tons per year.
The conventional means to control and minimize the glue application rate is to control the primary variables, the metering roll gaps, to empirical dimensional targets, to achieve approximate control of the top and bottom-side glue application rates. However, glue application rates are also affected by numerous secondary factors such as the medium""s physical properties (i.e. permeability), the glue""s characteristics (i.e. viscosity), the machine speed (which affects the amount of glue that is forced through a given metering roll gap), and the gap between the glue roll and the medium""s flute tips. An on-line, direct measurement of the glue application rate is needed to facilitate control of the glue application rate to optimum targets, using closed-loop feedback control of the metering roll gaps.
Machine speed on a corrugator varies frequently, often from as high as 800 feet per minute, down to 300 feet per minute, and back up, in less than a minute. Such rapid changes in speed produce concurrent, rapid changes in the glue application rate. To enable closed-loop control of the glue application rate throughout speed changes rapid measurement updates are needed, preferably every second or fraction of a second.
To date, no on-line, direct, rapid measurement of the glue application rate method has been successfully commercialized. U.S. Pat. No. 5,049,216 describes the basic principles of an infrared absorption sensor that with the proper selection of appropriate optical filters and light-sensitive detectors may be used to measure the adjacent quantities of certain organic materials, such as water, cellulose fiber, and starch. However, U.S. Pat. No. 5,663,565 indicates that the basic principle of infrared absorption cannot be used, without enhancement, to directly measure the amount of starch on the board, because the primary absorption bands for starch are the same as for cellulose fiber and because the water in the starch is indistinguishable from the water in the paper. U.S. Pat. No. 5,663,565 subsequently discloses an enhancement that allows the starch component of a direct infrared absorption measurement to be distinguished from the background measurement of the paper""s cellulose fiber and water, by exploiting the fact that the starch component of the measurement oscillates at the fluting frequency. However, the method proposed in U.S. Pat. No. 5,663,565 requires complex signal processing and a small light spot size that is tightly focused, depth-wise, on the flute tips, to prevent the undulating medium""s depth-sensitive fiber and water signatures from corrupting the measurement.
It is therefore a principle object of the present invention to provide a system and method for measuring the glue application rate on corrugated board.
It is a further object of the present invention to measure the glue application rate on corrugated board during the manufacture of such corrugated board.
Accordingly the system and method of the present invention uses common infrared sensor technology to successfully measure the rate at which glue is applied to corrugated board on a corrugator. This is achieved by first measuring the thickness of the aqueous glue film on the glue roll, both upstream and downstream of the point of application of the glue to the medium, using a pair of co-linear infrared absorption sensors, and then by subtracting the downstream value from the upstream one to derive the glue consumption or application rate.
The system and method of the present invention includes a first infrared absorption sensor that is located so as to measure the thickness of the aqueous glue film on the glue roll upstream of the where the glue film contacts the medium""s flutes. The point where the glue film contacts the medium""s flutes shall hereinafter be referred to as the gluing nip, and the upstream sensor shall hereinafter be referred to as the upstream sensor. The system and method of the present invention also includes a second infrared absorption sensor that is located so as to measure the thickness of the aqueous glue film on the glue roll just downstream of the gluing nip. This downstream sensor shall hereinafter be referred to as the downstream sensor.
The upstream and downstream sensors both measure the respective upstream and downstream films in the same manner. The preferred sensor design incorporates a light source to illuminate the glue film, a beam splitter to split the returning light into two paths, one band-pass filter for each light path to restrict each filtered light beam to a chosen narrow band-width, and one light-sensitive detector downstream of each band-pass filter to measure the intensity of the respective filtered light.
One band-pass filter is chosen to pass light in a narrow bandwidth that is heavily and uniquely absorbed by the water in the glue. The second band-pass filter is chosen to pass light within a second narrow bandwidth that is effectively not absorbed by either the water or the solids in the glue. The first detector, which shall hereinafter be referred to as the water-channel detector, produces an electrical output signal whose magnitude is inversely proportional to the amount of water in the glue film, which in turn is proportional to the glue film""s thickness. This first channel""s output signal is also generally proportional to the water-channel detector""s temperature that affects its signal gain, inversely proportional to the optical scattering effects of the glue film, and directly proportional to the intensity of the light source. The second detector, which shall hereinafter be referred to as the reference-channel detector, produces an electrical output signal whose magnitude is essentially only proportional to its temperature, inversely proportional to the optical scattering effects of the glue film, and directly proportional to the intensity of the light source.
The first step in the preferred signal processing method then computes the ratio of each sensor""s two detector output signals to derive a normalized, initial measurement value for each sensor. These two normalized, initial measurement values are thus essentially insensitive to changes in the internal temperature of each sensor, the scattering effects of the upstream and downstream glue films, and the intensity of the two sensor""s light sources. The resulting normalized, initial measurement values therefore effectively vary only with the amount of water in the upstream and downstream glue films, and hence only with the thicknesses of the upstream and downstream glue films.
Each sensor""s normalized, initial measurement value is then linearized by suitable means, such as by computing its logarithm, then multiplied by an empirically determined slope or gain, and finally added to an empirically determined offset value, to derive each sensor""s final calibrated measurement value. Suitable slopes and offsets are empirically determined by routine calibrations that compare manually measured glue film thicknesses to on-line sensor measurements.
The downstream sensor""s final calibrated measurement value is then subtracted from the upstream sensor""s final calibrated measurement value to derive the reduction in film thickness that results from the intervening application of glue to the medium""s flute tips. The computed reduction in glue film thickness is then multiplied by a constant that takes into account the assumed density and consistency (percent-solids) of the liquid glue to derive an intermediate solids consumption rate, defined as the weight of solid starch consumed per unit length of the glue roll""s circumference. The computed intermediate solids consumption rate is then multiplied by the measured rotational speed of the glue roll, in revolutions per unit time, and then by the known circumference of the glue roll, to compute the solids application rate, defined as the weight of solid starch applied to the medium""s flute tips per unit time. The measured machine speed is then multiplied by the known width of the manufactured corrugated board to compute the board production rate, defined as the surface area of board produced per unit time. The solids application rate is then multiplied by the computed board production rate to compute the final glue application rate, in weight of solid starch applied per unit surface area of board produced.
The final measured glue application rate may be derived using a fixed pair of co-linear sensors located at one measurement location, at a suitable position across the width of either the singlefacer or glue-machine, to provide a single measurement of the glue application rate versus time on just one side of the corrugated board. A second set of co-linear sensors may be located at another measurement location to derive a similar second fixed-point measurement of the glue application rate versus time on the other side of the corrugated board. At either measurement location an array of co-linear sensors across the width of the machine may alternatively be used to derive a discrete number of starch application measurements across the width of the singlefacer or glue-machine, so as to measure the cross-machine uniformity of the glue application. In still another arrangement, at any measurement location a pair of co-linear sensors may be mounted to a pair of synchronized scanning devices to mechanically convey the co-linear sensor pair back and forth across the width of the corrugating machine. This scanning arrangement therefore facilitates the derivation of essentially continuous glue application profiles across the width of the machine, to further improve the measurement of cross-machine glue uniformity.
These and other features and objects of the present invention will be more fully understood from the following detailed description, which should be read in light of the accompanying drawings in which corresponding reference numerals refer to corresponding parts through the several views.